Genetic factors play an important role in psoriatic arthritis (PsA), and both PsA and purely cutaneous psoriasis (PsC) display strong HLA associations. Through a combination of genomic resequencing and haplotype analysis, we have identified HLA-Cw6 as the major genetic determinant of psoriasis in Caucasians. However, substantial literature suggests that in addition to HLA-Cw6, there may be other MHC associations in PsA, particularly with HLA-B38 and -B39. In our 2003 submission, we described a 9.7 kb deletion located between HLA-B and HLA-C, developed an assay to type the deletion, and presented preliminary data suggesting that this deletion might account for PsA associations with both HLA-Cw6 and HLA-B38/39. However, our subsequent work in larger samples has shown that this deletion is unlikely to be causal. Instead, we and others have found evidence for HLA-Cw6-independent associations between PsA and MICA*002, a subtype of MICA-A9, and the TNF*-857T promoter polymorphism. Based on these results, we now hypothesize that there are two determinants of PsA susceptibility in the MHC, one in linkage disequilibrium (LD) with MICA*002, and one in LD with HLA-Cw6. To test this hypothesis, we have exceeded our recruiting goals for PsA, PsC, and normal controls, forged a collaboration with PsA experts Dafna Gladman and Proton Rahman that greatly increased our sample size, and generated a wealth of MHC SNP genotyping data in our recent genome-wide association scan (GWAS), the Collaborative Association Study of Psoriasis (CASP). In the context of the CASP replication study we are currently genotyping 1,800 MHC SNPs in 1,500 PsA cases, 1,500 PsC cases and 2,400 normal controls. With a view towards scanning the rest of the genome for PsA loci, we assessed the top 7 confirmed associations from CASP (HLA-C, IL12B, TNIP1, IL13, TNFAIP3, IL23A, and IL23R) in 1,755 PsA cases vs. 5,942 normal controls, finding that all of them except IL23R were associated with PsA. We also used the CASP data to perform a small-scale preliminary GWAS of PsA, subjecting 55 of the best signals to follow-up in 1,149 PsA cases and 2,301 controls. Eight signals replicated at p <0.05, and one of these was a possible PsA-specific allele at IL12B (rs4379175, r2 = 0.29 with the best PsC SNP, p = 3.3 x 10-8 discovery, 3.8 x 10-8 replication, 2.4 x 10-14 combined). While these preliminary results are intriguing and demonstrate our capabilities, the CASP PsA sample is too small to provide an adequately-powered GWAS. Therefore, to test our hypothesis of dual determinants of PsA in the MHC, and to expand our search for PsA genes to the rest of the human genome, we propose the following specific aims: 1. To accurately phenotype joint and skin involvement, to collect blood samples, and to prepare immortalized lymphoblastoid cell lines and DNA from at least 1,000 additional patients with PsA. 2. To perform a discovery GWAS of at least 1,800 PsA cases, 1150 PsC cases and 2,300 normal controls, taking advantage of existing GWAS data on 350 PsA cases, 900 PsC cases and 1,400 normal controls already available through CASP, and another 250 PsC cases and 300 normal controls to be scanned with the support of AR054966. 3. To follow up the non-MHC regions identified in Aim 2 on at least 2,000 PsA cases and 2,000 normal controls. Confirmed hits will be fine-mapped in our entire sample of PsA, PsC, and normal controls. 4. To use the results of aims 2 and 3 to identify MHC-encoded PsA determinants by recombinant ancestral haplotype analysis, following up implicated regions with additional genotyping and a search for shared HLA structural determinants. PUBLIC HEALTH RELEVANCE: Psoriatic arthritis is a painful and debilitating inflammatory disease of the skin and joints, affecting over 1 million Americans. Its cause is not well understood. Identification of genes that increase risk of psoriatic arthritis will help us unlock the secrets of this troubling disease, and identify targets for more specific and effective therapy.